CodeDump

My new favorite function: map

Ezra — Mon, 28 Dec 2009 03:46:55 +0000

Map, that’s my new favorite pure function. It’s very simple in use and has tons of uses! Ever been fidgeting with numbers trying to get the right output? Save your time the next time and use Map!

1
2
3
4

function map($v, $a, $b, $x = 0, $y = 1)
{
return ($v == $a) ? $x : ($v - $a) * ($y - $x) / ($b - $a) + $x;
}

As you can see it’s very easy in use. The first parameter is the value that needs to be mapped, the next two values are respectfully the minimum and maximum value of the input and the last two are respectfully the min and max of the output.

So lets take an example: It’s 20 degrees Celsius outside, but we want to map it to Fahrenheit.

1	map($_GET['c'], -40, 200, -40, 392);

And that’s it!, we know that -40 degrees Celsius is -40 degrees Fahrenheit and that 200 degrees Celsius is 392 degrees Fahrenheit, and this way the value of c gets mapped from Celsius to Fahrenheit, that simple! No more fidgeting with formulae.

Also because it’s a pure function it’s easily memoized, this will make the already fast function even faster by caching the inputs and the outputs. But because I don’t know an easy way of Memoizing functions in PHP except for explicitly making the function memoized, i’ll leave it with a link to Wikipedia.

This is a post from CodeDump

My new favorite function: map

Firefox and the Cache Mystery

Ezra — Wed, 15 Jul 2009 20:08:17 +0000

Lately I’ve been working on a small and fast CMS system for my new website. I’m hoping the website will attract a lot of visitors and since my server is not that strong I’ve decided on sending some caching headers so the browsers won’t overwhelm the server with requests.

After some research the Cache-Control header will work the best for me. It sets an ‘freshness’ time to the resource and during this time the browser won’t request the resource again and simply load it from disk.

This might seem like a very easy case but it did give me some problems. After sending the headers I verified with firebug that the header was being sent, but when I refreshed the page, Firefox was requesting the resource again. I was stunned, here I was sending the right headers and somehow the browser was simply ignoring me… or was it?

After some asking around I was still getting nowhere, somehow Firefox just did not want to cache my resources. But after much searching around on the internet, I finally found my answer. Firefox does not load resources from the cache when refreshing!.

I don’t know why, but I guess it makes sense somehow. But I don’t know why they didn’t move this to the Hard-Refresh. (Ctrl+F5 or Ctrl+Shift+R). If anyone can answer me this, please do.

This is a post from CodeDump

Firefox and the Cache Mystery

OpenSearch Wordpress Plugin

Ezra — Mon, 06 Oct 2008 18:01:47 +0000

I’ve been searching for a plugin for Wordpress that makes an OpenSearch file.

Found nothing so i’ve been working on it myself, however lately I’ve seen another OpenSearch plugin appear in the plugins database.

The plugin system of wordpress is ok, but could be a lot better, namely a class system. The way it works now is there are a lot of events that you can register functions to, that will fire when the event happens. You can also create some filters that will for example pass the text that has been written in a new post to alter some text like tags added to the post.

It would have been very nice if the plugin system would have worked with a class system, by simply creating a class to expand on the features.

But enough talk about the plugin system, let me tell you some about the plugin. It’s my first plugin so I’m still experimenting with the API. The way it works now is it creates a OpenSearch.xml file when the plugin activates and the file is basically static with some variable values that make the file specific for your blog.

As I said before, now someone else started working on the plugin so mine has become useless as the other one is superior to mine. So without further ado, the code that I had for anyone to download if they please.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110

/**
Edits:
Ezra 18-08-08 Eerste versie
Ezra 19-08-08 Hij werkt, gzip ook!
Ezra 28-08-08 Bij Deactivaten wordt het bestand verwijderd. Encoding veranderd naar US-ASCII.

*/

/*
Plugin Name: OpenSearch
Plugin URI: http://codedump.tsdme.nl
Description: Add a OpenSearch description to your blog
Version: 0.1
Author: Ezra (Ezra@tsdme.nl)
Author URI: http://codedump.tsdme.nl/Admin
*/

//Definitions
define("OS_LOCATION", "OpenSearch.xml");
define("PLUGIN_BORDER", "\t\n");

//Hooks
register_activation_hook(__FILE__, 'OS_Activate');
register_deactivation_hook(__FILE__, 'OS_Deactivate');
add_action('wp_head', OS_Head, 10, 0);

//Functions
function SO_add_pages(){
add_options_page('OpenSearch', 'OpenSearch', 10, __FILE__, 'OS_Menu');
}
function OS_Activate(){
clearstatcache();

$os_file = "\"1.0\" encoding=\"US-ASCII\"?>\n";
$os_file .= "\"http://a9.com/-/spec/opensearch/1.1/\">\n";
$os_file .= "\t".get_bloginfo('name')."\n";
$os_file .= "\tSearch ".get_bloginfo('name')."\n";
$os_file .= "\tSearch ".get_bloginfo('home')."\n";
$os_file .= "\t".get_the_tag_list('',',','')."\n";
$os_file .= "\t".get_bloginfo('admin_email')."\n";
$os_file .= "\t\"text/html\" template=\"http://gedachtegoed.tsdme.nl/?s={searchTerms}\"/>\n";
$os_file .= "\t\"48\" width=\"48\" type=\"image/png\">".get_bloginfo('home')."/favicon.png\n";
$os_file .= "\t\"example\" searchTerms=\"cat\" />\n";
$os_file .= "";

$os_xml = @fopen(GetHomePath().OS_LOCATION, "wb");
if($os_xml){
flock($os_xml, LOCK_EX);
fwrite($os_xml, $os_file);
fclose($os_xml);
}
$gz_xml = @gzopen(GetHomePath().OS_LOCATION.".gz", "wb9");
if($gz_xml){
flock($gz_xml, LOCK_EX);
gzwrite($gz_xml, $os_file);
gzclose($gz_xml);
}
}

function OS_Deactivate(){
if(file_exists(GetHomePath().OS_LOCATION)){
unlink(GetHomePath().OS_LOCATION);
}
if(file_exists(GetHomePath().OS_LOCATION.".gz")){
unlink(GetHomePath().OS_LOCATION.".gz");
}
}

function OS_Head(){
echo PLUGIN_BORDER;
echo "\t\"search\" type=\"application/opensearchdescription+xml\" title=\"".get_bloginfo('name')." Search\" href=\"".get_bloginfo('url')."/".OS_LOCATION."\" />\n";
echo PLUGIN_BORDER;
}

/**
* Returns the path to the blog directory
*
* @since 3.0
* @access public
* @author Arne Brachhold
* @return string The full path to the blog directory
*/
function GetHomePath() {

$res="";
//Check if we are in the admin area -> get_home_path() is avaiable
if(function_exists("get_home_path")) {
$res = get_home_path();
} else {
//get_home_path() is not available, but we can't include the admin
//libraries because many plugins check for the "check_admin_referer"
//function to detect if you are on an admin page. So we have to copy
//the get_home_path function in our own...
$home = get_option( 'home' );
if ( $home != '' && $home != get_option( 'siteurl' ) ) {
$home_path = parse_url( $home );
$home_path = $home_path['path'];
$root = str_replace( $_SERVER["PHP_SELF"], '', $_SERVER["SCRIPT_FILENAME"] );
$home_path = trailingslashit( $root.$home_path );
} else {
$home_path = ABSPATH;
}

$res = $home_path;
}
return $res;
}
?>

Have fun with it and good luck to the person that is developing the other OpenSearch plugin!

This is a post from CodeDump

OpenSearch Wordpress Plugin

Cookies and Actionscript

Ezra — Thu, 10 Jul 2008 04:22:30 +0000

I have been searching and searching today to find out a way to see the cookies that the server sends to you while downloading a page using URLLoader. I needed to read the PHPSESSID cookie that php send me so I could use it again in another URLLoader object so I could download some information that was protected behind a login. With one URLLoader object I sent the necessary form information to the login script and then I would transfer the PHPSESSID cookie to new URLLoader objects to download the information I wanted that was stored on different pages while still being authenticated with the correct session ID.

Sadly tho, after searching long I could not find anything that would tell me how to read the cookies from the URLLoader object or how to transfer them to the new objects. –I found a lot about the Flash Shared Object, but that was not what I was looking for, sadly–

I’m still not sure why, but for some reason I ran the code without copying the session ID to the new object to see what would happen and waddayaknow, I was getting the information I wanted. So somehow cookies are stored automatically from one URLLoader object to another, what struck me as strange, because I thought I would be starting with a clean sheet when I created a new object, not that cookies –and who knows what else– would be stored between them. So next time I will have to find out how to open multiple sessions at the same time, but I’ll do that when I’ll need to.

If anyone does know how to see what cookies were sent and what their value is, please let me know as I’m sure I could use it sometime.

P.S: For anyone wondering what I was making. My university has a special page with all my grades and ECTS credits and I wanted an easy way of checking them and maybe alerting me when the page was updated. So I decided to create it with AIR because some of my classmates –That might find it handy too– have macs while I use a pc, so it would be OS Independent.

This is a post from CodeDump

Cookies and Actionscript

Conway’s Game of Life

Ezra — Fri, 27 Jun 2008 22:52:23 +0000

Every time I try to learn a new programming language as a ‘Hello World’ world program I write Conway’s Game of Life.

So when John Resig created processing.js I started building again.

I’m not sure why, but the program is using a LOT of cpu cycles, I tried to optimize as much as I can, but i’m not sure if it worked as they were javascript optimization and I don’t know if they apply to processing too. Also I found out that some things still don’t work, most processing functions are working natively in javascript but some also don’t exist and are not put into processing.js yet. Like the function round, that like the name implies rounds the number to it’s closest integer value. I tried to submit a bug at the github page for processing but there is no such thing. The way I solved my problem is by simply adding 1 to the number and using the floor function that is implemented.

I’ve tried numerous ways of optimizing the program and been timing everything with the awesome add-on Firebug. The test results are noted in the code and if anyone knows any ways to optimize it better please let me know.

But let’s not ramble on too much about the code and just show it.

/** A little test to see if processing.js is really that good...
- So, not to break with tradition I wrote my usual Hello World
- Conway's Game of Life
-
- It has a couple of optimizations, namely calculating the new
- matrix and visualizing them is done at the same time.
- The matrix consists of a ring of 'death'cells as a safe sandbox
- to avoid memory errors
- The 2 matrices are not copied over after calculating, but they
- are swapped.
- Loops are run in reverse because it's faster to check a value to
- zero than to another value (w&h).

Author: Ezra Pool (ezra@tsdme.nl);
Version: 0.4
*/
int a,b,w,h;
int[][][] m;

void setup() {
h=250;w=250;a=0;b=1;
m = new int[2][w+1][h+1];

size(w, h);
frameRate(5);

spawn();

}
void draw() {

background(255);
lifedeathandshow();

}
void spawn(){
int i,j;

/**
- Apparantly it's faster to run a loop in reverse because checking to see
- if a value is zero is faster than checking another value (w & h).
-
- Testing results:
- normalloop: 148ms
- reverseloop: 145ms
//*/
for(i=1;i0;i--){
for(j=h-1;j>0;j--){
m[a][i][j] = floor(random(0,2));
}
}
}

void lifedeathandshow(){
int c,i,j,n;

/**
- Apparantly it's faster to run a loop in reverse because checking to see
- if a value is zero is faster than checking another value (w & h).
-
- Testing results don't show much improvement, some rounds are faster
- some are slower, but this is mostly due to the complexity of some of
- the iterations. (More alive or dead cells).

for(i=1;i< 2 || n > 3){
m[b][i][j] = 0;
}else{
m[b][i][j] = m[a][i][j];
point(i,j);
}
}else{
if(n == 3){
m[b][i][j] = 1;
point(i,j);
}else{
m[b][i][j] = m[a][i][j];
}
}
}
}
//*/

for(i=w-1;i>0;i--){
for(j=h-1;j>0;j--){
n = neighbours(i,j);
if(m[a][i][j] == 1){
if(n < 2 || n > 3){
m[b][i][j] = 0;
}else{
m[b][i][j] = m[a][i][j];
point(i,j);
}
}else{
if(n == 3){
m[b][i][j] = 1;
point(i,j);
}else{
m[b][i][j] = m[a][i][j];
}
}
}
}

/** I don't know which of the two versions is faster in javascript/processing...
- is the temp variable one faster or the XOR swap??
-
- Testing with Firebug in Mozilla/5.0 Gecko/2008062505 GranParadiso/3.0.1pre
- Results
- tempvar: 171ms (10.000 iterations)
- xorswap: 203ms (10.000 iterations)
*/

c=b;
b=a;
a=c;

//a ^= b;
//b ^= a;
//a ^= b;
}
int neighbours(int i, int j){
return m[a][i-1][j-1] +
m[a][i][j-1] +
m[a][i+1][j-1] +
m[a][i-1][j] +
m[a][i][j] +
m[a][i+1][j] +
m[a][i-1][j+1] +
m[a][i][j+1] +
m[a][i+1][j+1];
}

This is a post from CodeDump

Conway’s Game of Life